package neural_nets_lib

  1. Overview
  2. Docs
A from-scratch Deep Learning framework with an optimizing compiler, shape inference, concise syntax

Install 

dune-project
 Dependency

github.com Readme Changelog Edit opam file Versions (6)

Authors

  1. L
    Lukasz Stafiniak

Maintainers

  1. L
    Lukasz Stafiniak <lukstafi@gmail.com>

Sources

0.3.3.3.tar.gz
md5=9170d4d98422350c9a73a95adfb795dc
sha512=c1b024a69b1d0338af6e34508dbf6dccf3c2b6cc156e7628c3d7853c7040e225bdfc0a8731bb4db5a97edba90e26439987bfa505154d23af46f119c07ad809ed

doc/neural_nets_lib/Ocannl/Operation/TDSL/index.html

Module Operation.TDSLSource

include module type of struct include Initial_TDSL end
Sourceval term : label:Base.string Base.list -> ?batch_dims:Base.int Base.list -> ?input_dims:Base.int Base.list -> ?output_dims:Base.int Base.list -> ?batch_axes:(Base.string * Base.int) Base.list -> ?input_axes:(Base.string * Base.int) Base.list -> ?output_axes:(Base.string * Base.int) Base.list -> ?deduced:Shape.deduce_within_shape -> ?init_op:Tensor.init_op -> ?fetch_op:(v:Tensor.tn -> Tensor.fetch_op) -> Base.unit -> Tensor.t
Sourceval number : ?label:Base.string Base.list -> ?axis_label:Base.string -> Base.float -> Tensor.t
Sourceval ndarray : ?label:Base.string Base.list -> ?batch_dims:Base.int Base.list -> ?input_dims:Base.int Base.list -> ?output_dims:Base.int Base.list -> ?batch_axes:(Base.string * Base.int) Base.list -> ?input_axes:(Base.string * Base.int) Base.list -> ?output_axes:(Base.string * Base.int) Base.list -> ?strict:Base.bool -> Base.float Base.array -> Tensor.t
Sourceval param : ?input_dims:Base.int Base.list -> ?output_dims:Base.int Base.list -> ?input_axes:(Base.string * Base.int) Base.list -> ?output_axes:(Base.string * Base.int) Base.list -> ?deduced:Shape.deduce_within_shape -> ?strict:Base.bool -> ?values:Base.float Base.array -> Base.string -> Tensor.t
Sourcemodule O = DO
Sourceval einsum : ?label:Base.string list -> Base.string -> Tensor.t -> Tensor.t -> Tensor.t
Sourceval outer_sum : ?label:Base.string list -> Base.string -> Tensor.t -> Tensor.t -> Tensor.t
Sourceval einsum1 : ?label:Base.string list -> Base.string -> Tensor.t -> Tensor.t
Sourceval range : ?label:Base.string list -> ?axis_label:Base.string -> Base.Int.t -> Tensor.t
Sourceval range_of_shape : ?label:Base.string list -> ?batch_dims:Base.Int.t Base.List.t -> ?input_dims:Base.Int.t Base.List.t -> ?output_dims:Base.Int.t Base.List.t -> ?batch_axes:(Base.string * Base.Int.t) Base.List.t -> ?input_axes:(Base.string * Base.Int.t) Base.List.t -> ?output_axes:(Base.string * Base.Int.t) Base.List.t -> unit -> Tensor.t
Sourceval stop_gradient : ?label:Base.string list -> Tensor.t -> Tensor.t

The input i dimensions default to empty. The batch dimensions will be inferred if omitted. strict controls whether Constant_fill will try to fit the given values in the tensor and contribute to shape inference. If it is not provided explicitly, it will be true if b is omitted, and false otherwise.

It's like `Tensor.param` but without shape inference.